init update

Voldemort108X · Jul 30, 2021 · c3fbe88 · c3fbe88
1 parent 6ae88e8
commit c3fbe88
Show file tree

Hide file tree

Showing 4 changed files with 341 additions and 27 deletions.
diff --git a/README.md b/README.md
@@ -16,41 +16,78 @@
 }
 ```
 ## Dataset
-Please refer to challenge website [[link]](https://zmiclab.github.io/projects/myops20/) for dataset access.
+Please refer to challenge website [[link]](https://zmiclab.github.io/projects/myops20/) for dataset access. The dataset contains three folders: train25, train25_myops_gd, test20.
 
 ## Environment
 1. UNet: 
-
+```
+conda env create -f myops_unet.yml
+```
 2. Mask-RCNN and UNet++:
+```
+conda env create -f myops_mrcnn_unetpp.yml
+```
+## Default directory structure
 
-## Default files
+    ├── Data                   
+    |   ├── Original_data       # Place the downloaded dataset here
+    |   |   ├── train25
+    |   |   ├── train25_myops_gd
+    |   |   ├── test20
+    ├── mask_rcnn_coco.h5       # Downloaded pre-trained mask_rcnn weights
+    ├── config.py
+    ├── data_creator.py
+    ├── ...
 
 
 ## Setup
-1. Preprocessing
-    - Extract dataset to 2D slices
+1. Data creator including random warping augmentation
+```
+python data_creator.py 
+```
+
+2. Train networks
+    - Train UNet for LV_BP, RV_BP, LV_NM, LV_ME, LV_MS blocks
     ```
-    python data_creator.py
+    python train_UNet.py
+    ```
+    - Train Mask-RCNN for LV_ME and LV_MS blocks
+        - Download pretrained mask_rcnn_coco.h5 at [[here]](https://github.com/matterport/Mask_RCNN/releases) and place it in the current folder.
+        - Train Mask-RCNN for LV_ME block
+        ```
+        python train_MaskRCNN.py --mode 'LV_ME'
+        ```
+        - Train Mask-RCNN for LV_MS block
+        ```
+        python train_MaskRCNN.py --mode 'LV_MS'
+        ```
+    - Train UNet++ for LV_ME and LV_MS blocks
+    ```
+    python train_UNetplusplus.py
     ```
-
-2. Train networks
-    - Train UNet for
-    - Train Mask-RCNN
-        - Download mrcnn_coco.h5 
-        - 
-    - Train UNet++ for 
-
 3. Test networks:
     - Test UNet
+    ```
+    python test_UNet.py
+    ```
     - Test Mask-RCNN
-    - Test UNet++ 
-
-4. Post-processing and linear decoder:
-    - Post processing
+        - Test LV_ME block
+        ```
+        python test_MaskRCNN.py --mode 'LV_ME'
+        ```
+        - Test LV_MS block
+        ```
+        python test_MaskRCNN.py --mode 'LV_MS'
+        ```
+    - Test UNet++
     ```
-    python post_processing.py
+    python test_UNetplusplus.py
     ```
 
+4. Post-processing and linear decoder:
+```
+python post_processing.py
+```
 ## Acknowledgement
 1. Please cite the official Mask-RCNN and UNet++ implementations if you use them:
     - Mask-RCNN: https://github.com/matterport/Mask_RCNN

diff --git a/helper.py b/helper.py
@@ -235,29 +235,21 @@ def func_saveTrainingDataIntoSlices(destPath, train_inputs, train_labels, mode):
         np.save(os.path.join(destPath_label, str(i+1)+'.npy'), train_labels[i])
 
 def func_createLabelToMask(label,myo_type):
-
-    #print(type(label))
-    #print(label.shape)
     size_x,size_y = label.shape
     mask = np.zeros(shape=(size_x,size_y))
 
-        #indices = [(row,col.index(class_values[i])) for row, col in enumerate(list(label)) if class_values[i] in row]
     if myo_type == 'LV_MS' or myo_type == 'BG':
         indices = np.where(label==class_values_mrcnn_dict[myo_type])
         x = indices[0]
         y = indices[1]
         for j in range(indices[0].shape[0]):
-                #print(x[j],y[j])
             mask[(x[j],y[j])] = 1
 
         return mask
 
     if myo_type == 'LV_ME':
         indices = np.where(label==class_values_mrcnn_dict[myo_type])
         indices_LVMS = np.where(label==LV_MS)
-        #print(type(indices[1]))
-        #print(len(indices))
-        #print(indices[0].shape)
         x = indices[0]
         y = indices[1]
         for j in range(indices[0].shape[0]):

diff --git a/myops_mrcnn_unetpp.yml b/myops_mrcnn_unetpp.yml
@@ -0,0 +1,139 @@
+name: myo20_mrcnn
+channels:
+- defaults
+dependencies:
+- _libgcc_mutex=0.1=main
+- _tflow_select=2.1.0=gpu
+- absl-py=0.9.0=py36_0
+- astor=0.8.0=py36_0
+- backcall=0.2.0=py_0
+- blas=1.0=mkl
+- bzip2=1.0.8=h7b6447c_0
+- c-ares=1.15.0=h7b6447c_1001
+- ca-certificates=2020.6.24=0
+- cairo=1.14.12=h8948797_3
+- certifi=2020.6.20=py36_0
+- cloudpickle=1.5.0=py_0
+- cudatoolkit=10.0.130=0
+- cudnn=7.6.5=cuda10.0_0
+- cupti=10.0.130=0
+- cycler=0.10.0=py36_0
+- cytoolz=0.10.1=py36h7b6447c_0
+- dask-core=2.20.0=py_0
+- dbus=1.13.16=hb2f20db_0
+- decorator=4.4.2=py_0
+- expat=2.2.9=he6710b0_2
+- ffmpeg=4.0=hcdf2ecd_0
+- fontconfig=2.13.0=h9420a91_0
+- freeglut=3.0.0=hf484d3e_5
+- freetype=2.10.2=h5ab3b9f_0
+- gast=0.2.2=py36_0
+- glib=2.65.0=h3eb4bd4_0
+- google-pasta=0.2.0=py_0
+- graphite2=1.3.14=h23475e2_0
+- grpcio=1.27.2=py36hf8bcb03_0
+- gst-plugins-base=1.14.0=hbbd80ab_1
+- gstreamer=1.14.0=hb31296c_0
+- h5py=2.8.0=py36h61e79e4_3
+- harfbuzz=1.8.8=hffaf4a1_0
+- hdf5=1.8.20=hba1933b_1
+- icu=58.2=he6710b0_3
+- imageio=2.9.0=py_0
+- intel-openmp=2020.1=217
+- ipykernel=5.3.4=py36h5ca1d4c_0
+- ipython=7.16.1=py36h5ca1d4c_0
+- ipython_genutils=0.2.0=py36_0
+- jasper=2.0.14=h07fcdf6_1
+- jedi=0.17.1=py36_0
+- joblib=1.0.1=pyhd3eb1b0_0
+- jpeg=9b=h024ee3a_2
+- jupyter_client=6.1.12=pyhd3eb1b0_0
+- jupyter_core=4.7.1=py36h06a4308_0
+- keras=2.3.1=0
+- keras-applications=1.0.8=py_1
+- keras-base=2.3.1=py36_0
+- keras-preprocessing=1.1.0=py_1
+- kiwisolver=1.2.0=py36hfd86e86_0
+- lcms2=2.11=h396b838_0
+- ld_impl_linux-64=2.33.1=h53a641e_7
+- libedit=3.1.20191231=h14c3975_1
+- libffi=3.3=he6710b0_2
+- libgcc-ng=9.1.0=hdf63c60_0
+- libgfortran-ng=7.3.0=hdf63c60_0
+- libglu=9.0.0=hf484d3e_1
+- libopencv=3.4.2=h765d7f9_1
+- libopus=1.3.1=h7b6447c_0
+- libpng=1.6.37=hbc83047_0
+- libprotobuf=3.12.3=hd408876_0
+- libsodium=1.0.18=h7b6447c_0
+- libstdcxx-ng=9.1.0=hdf63c60_0
+- libtiff=4.1.0=h2733197_1
+- libuuid=1.0.3=h1bed415_2
+- libvpx=1.7.0=h439df22_0
+- libxcb=1.14=h7b6447c_0
+- libxml2=2.9.10=he19cac6_1
+- lz4-c=1.9.2=he6710b0_0
+- markdown=3.1.1=py36_0
+- matplotlib=3.2.2=0
+- matplotlib-base=3.2.2=py36hef1b27d_0
+- mkl=2020.1=217
+- mkl-service=2.3.0=py36he904b0f_0
+- mkl_fft=1.1.0=py36h23d657b_0
+- mkl_random=1.1.1=py36h0573a6f_0
+- ncurses=6.2=he6710b0_1
+- networkx=2.4=py_1
+- numpy=1.18.5=py36ha1c710e_0
+- numpy-base=1.18.5=py36hde5b4d6_0
+- olefile=0.46=py_0
+- opencv=3.4.2=py36h40b0b35_1
+- openssl=1.1.1g=h7b6447c_0
+- opt_einsum=3.1.0=py_0
+- parso=0.7.0=py_0
+- pcre=8.44=he6710b0_0
+- pexpect=4.8.0=py36_0
+- pickleshare=0.7.5=py36_0
+- pillow=7.2.0=py36hb39fc2d_0
+- pip=20.1.1=py36_1
+- pixman=0.40.0=h7b6447c_0
+- prompt-toolkit=3.0.5=py_0
+- protobuf=3.12.3=py36he6710b0_0
+- ptyprocess=0.6.0=py36_0
+- py-opencv=3.4.2=py36h765d7f9_1
+- pygments=2.6.1=py_0
+- pyparsing=2.4.7=py_0
+- pyqt=5.9.2=py36h05f1152_2
+- python=3.6.10=h7579374_2
+- python-dateutil=2.8.1=py_0
+- pywavelets=1.1.1=py36h7b6447c_0
+- pyyaml=5.3.1=py36h7b6447c_1
+- pyzmq=20.0.0=py36h2531618_1
+- qt=5.9.7=h5867ecd_1
+- readline=8.0=h7b6447c_0
+- scikit-image=0.16.2=py36h0573a6f_0
+- scipy=1.5.0=py36h0b6359f_0
+- setuptools=49.2.0=py36_0
+- sip=4.19.8=py36hf484d3e_0
+- six=1.15.0=py_0
+- sqlite=3.32.3=h62c20be_0
+- tensorboard=1.15.0=pyhb230dea_0
+- tensorflow=1.15.0=gpu_py36h5a509aa_0
+- tensorflow-base=1.15.0=gpu_py36h9dcbed7_0
+- tensorflow-estimator=1.15.1=pyh2649769_0
+- tensorflow-gpu=1.15.0=h0d30ee6_0
+- termcolor=1.1.0=py36_1
+- tk=8.6.10=hbc83047_0
+- toolz=0.10.0=py_0
+- tornado=6.0.4=py36h7b6447c_1
+- traitlets=4.3.3=py36_0
+- wcwidth=0.2.5=py_0
+- webencodings=0.5.1=py36_1
+- werkzeug=0.16.1=py_0
+- wheel=0.34.2=py36_0
+- wrapt=1.12.1=py36h7b6447c_1
+- xz=5.2.5=h7b6447c_0
+- yaml=0.2.5=h7b6447c_0
+- zeromq=4.3.4=h2531618_0
+- zlib=1.2.11=h7b6447c_3
+- zstd=1.4.4=h0b5b093_3
+prefix: /home/xiaoran8/.local/easybuild/software/2017/Core/miniconda3/4.3.27/envs/myo20_mrcnn
+